pygpt-net 2.4.36.post1__tar.gz → 2.4.38__tar.gz

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/CHANGELOG.md

@@ -1,5 +1,18 @@
 # CHANGELOG
 
+## 2.4.38 (2024-12-08)
+
+- Added the ability to select a style for chat display between: Blocks, ChatGPT-like, and ChatGPT-like Wide. New option in the menu: Config -> Theme -> Style...
+- Added configuration options for audio input in Settings -> Audio -> Audio Input Device, Channels, and Sampling rate.
+
+## 2.4.37 (2024-11-30)
+
+- The `Query only` mode in `Uploaded` tab has been renamed to `RAG`.
+- New options have been added under `Settings -> Files and Attachments`: 
+  - `Use history in RAG query`: When enabled, the content of the entire conversation will be used when preparing a query if the mode is set to RAG or Summary.
+  - `RAG limit`: This option is applicable only if 'Use history in RAG query' is enabled. It specifies the limit on how many recent entries in the conversation will be used when generating a query for RAG. A value of 0 indicates no limit.
+- Cache: dynamic parts of the system prompt (from plugins) have been moved to the very end of the prompt stack to enable the use of prompt cache mechanisms in OpenAI.
+
 ## 2.4.36 (2024-11-28)
 
 - Added a new command-line argument: --workdir="/path/to/workdir" to explicitly set the current working directory.
@@ -33,7 +46,7 @@
 
 - Added an option checkbox `Auto-index on upload` in the `Attachments` tab: 
 
-**Tip:** To use the `Query only` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `Query only` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
+**Tip:** To use the `RAG` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `RAG` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
 
 - Added context menu options in `Uploaded attachments` tab: `Open`, `Open Source directory` and `Open Storage directory`.
 

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pygpt-net
-Version: 2.4.36.post1
+Version: 2.4.38
 Summary: Desktop AI Assistant powered by models: OpenAI o1, GPT-4o, GPT-4, GPT-4 Vision, GPT-3.5, DALL-E 3, Llama 3, Mistral, Gemini, Claude, Bielik, and other models supported by Langchain, Llama Index, and Ollama. Features include chatbot, text completion, image generation, vision analysis, speech-to-text, internet access, file handling, command execution and more.
 Home-page: https://pygpt.net
 License: MIT
@@ -92,7 +92,7 @@ Description-Content-Type: text/markdown
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.4.36** | build: **2024.11.28** | Python: **>=3.10, <3.12**
+Release: **2.4.38** | build: **2024.12.08** | Python: **>=3.10, <3.12**
 
 > Official website: https://pygpt.net | Documentation: https://pygpt.readthedocs.io
 > 
@@ -693,11 +693,11 @@ Built-in file loaders:
 - Webpages (crawling any webpage content)
 - YouTube (transcriptions)
 
-You can configure data loaders in `Settings / LlamaIndex / Data Loaders` by providing list of keyword arguments for specified loaders.
+You can configure data loaders in `Settings / Indexes (LlamaIndex) / Data Loaders` by providing list of keyword arguments for specified loaders.
 You can also develop and provide your own custom loader and register it within the application.
 
 LlamaIndex is also integrated with context database - you can use data from database (your context history) as additional context in discussion. 
-Options for indexing existing context history or enabling real-time indexing new ones (from database) are available in `Settings / LlamaIndex` section.
+Options for indexing existing context history or enabling real-time indexing new ones (from database) are available in `Settings / Indexes (LlamaIndex)` section.
 
 **WARNING:** remember that when indexing content, API calls to the embedding model are used. Each indexing consumes additional tokens. Always control the number of tokens used on the OpenAI page.
 
@@ -759,7 +759,7 @@ You can set the limit of steps in such a loop by going to `Settings -> Agents an
 
 You can change the prompt used for evaluating the response in `Settings -> Prompts -> Agent: evaluation prompt in loop`. Here, you can adjust it to suit your needs, for example, by defining more or less critical feedback for the responses received.
 
-##  Agent (Legacy, Autonomous) 
+##  Agent (Autonomous) 
 
 This is an older version of the Agent mode, still available as legacy. However, it is recommended to use the newer mode: `Agent (LlamaIndex)`.
 
@@ -907,11 +907,13 @@ The content from the uploaded attachments will be used in the current conversati
 
 - `Full context`: Provides best results. This mode attaches the entire content of the read file to the user's prompt. This process happens in the background and may require a large number of tokens if you uploaded extensive content.
 
-- `Query only`: The indexed attachment will only be queried in real-time using LlamaIndex. This operation does not require any additional tokens, but it may not provide access to the full content of the file 1:1.
+- `RAG`: The indexed attachment will only be queried in real-time using LlamaIndex. This operation does not require any additional tokens, but it may not provide access to the full content of the file 1:1.
 
 - `Summary`: When queried, an additional query will be generated in the background and executed by a separate model to summarize the content of the attachment and return the required information to the main model. You can change the model used for summarization in the settings under the `Files and attachments` section.
 
-**Important**: When using `Full context` mode, the entire content of the file is included in the prompt, which can result in high token usage each time. If you want to reduce the number of tokens used, instead use the `Query only` option, which will only query the indexed attachment in the vector database to provide additional context.
+In the `RAG` and `Summary` mode, you can enable an additional setting by going to `Settings -> Files and attachments -> Use history in RAG query`. This allows for better preparation of queries for RAG. When this option is turned on, the entire conversation context is considered, rather than just the user's last query. This allows for better searching of the index for additional context. In the `RAG limit` option, you can set a limit on how many recent entries in a discussion should be considered (`0 = no limit, default: 3`).
+
+**Important**: When using `Full context` mode, the entire content of the file is included in the prompt, which can result in high token usage each time. If you want to reduce the number of tokens used, instead use the `RAG` option, which will only query the indexed attachment in the vector database to provide additional context.
 
 **Images as Additional Context**
 
@@ -919,7 +921,7 @@ Files such as jpg, png, and similar images are a special case. By default, image
 
 **Uploading larger files and auto-index**
 
-To use the `Query only` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `Query only` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
+To use the `RAG` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `RAG` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
 
 ## Downloading files
 
@@ -2768,6 +2770,8 @@ Config -> Settings...
 
 - `Zoom`: Adjusts the zoom in chat window (web render view). `WebEngine / Chromium` render mode only.
 
+- `Style (chat)`: Chat style (Blocks, or ChatGPT-like, or ChatGPT-like Wide. `WebEngine / Chromium` render mode only.
+
 - `Code syntax highlight`: Syntax highlight theme in code blocks. `WebEngine / Chromium` render mode only.
 
 - `Font Size (chat window)`: Adjusts the font size in the chat window (plain-text) and notepads.
@@ -2790,8 +2794,6 @@ Config -> Settings...
 
 - `Use theme colors in chat window`: Use color theme in chat window, Default: True.
 
-- `Disable markdown formatting in output`: Enables plain-text display in output window, Default: False.
-
 **Files and attachments**
 
 - `Store attachments in the workdir upload directory`: Enable to store a local copy of uploaded attachments for future use. Default: True
@@ -2800,6 +2802,16 @@ Config -> Settings...
 
 - `Directory for file downloads`: Subdirectory for downloaded files, e.g. in Assistants mode, inside "data". Default: "download"
 
+- `Verbose mode`: Enabled verbose mode when using attachment as additional context.
+
+- `Model for querying index`: Model to use for preparing query and querying the index when the RAG option is selected.
+
+- `Model for attachment content summary`: Model to use when generating a summary for the content of a file when the Summary option is selected.
+
+- `Use history in RAG query`: When enabled, the content of the entire conversation will be used when preparing a query if mode is RAG or Summary.
+
+- `RAG limit`: Only if the option `Use history in RAG query` is enabled. Specify the limit of how many recent entries in the conversation will be used when generating a query for RAG. 0 = no limit.
+
 **Context**
 
 - `Context Threshold`: Sets the number of tokens reserved for the model to respond to the next prompt.
@@ -2892,14 +2904,22 @@ Config -> Settings...
 
 **Vision**
 
+- `Vision: Camera Input Device`: Video capture camera index (index of the camera, default: 0).
+
 - `Vision: Camera capture width (px)`: Video capture resolution (width).
 
 - `Vision: Camera capture height (px)`: Video capture resolution (height).
 
-- `Vision: Camera IDX (number)`: Video capture camera index (number of camera).
-
 - `Vision: Image capture quality`: Video capture image JPEG quality (%).
 
+**Audio**
+
+- `Audio Input Device`: Selects the audio device for Microphone input.
+
+- `Channels`: Input channels, default: 1
+
+- `Sampling Rate`: Sampling rate, default: 44100
+
 **Indexes (LlamaIndex)**
 
 - `Indexes`: List of created indexes.
@@ -3337,7 +3357,7 @@ If you want to only query index (without chat) you can enable `Query index only
 
 You can create a custom vector store provider or data loader for your data and develop a custom launcher for the application. 
 
-See the section `Extending PyGPT / Adding custom Vector Store provider` for more details.
+See the section `Extending PyGPT / Adding a custom Vector Store provider` for more details.
 
 # Updates
 
@@ -3635,6 +3655,8 @@ Syntax: `event name` - triggered on, `event data` *(data type)*:
 
 - `AI_NAME` - when preparing an AI name, `data['value']` *(string, name of the AI assistant)*
 
+- `AGENT_PROMPT` - on agent prompt in eval mode, `data['value']` *(string, prompt)*
+
 - `AUDIO_INPUT_RECORD_START` - start audio input recording
 
 - `AUDIO_INPUT_RECORD_STOP` -  stop audio input recording
@@ -3693,10 +3715,16 @@ Syntax: `event name` - triggered on, `event data` *(data type)*:
 
 - `POST_PROMPT` - after preparing a system prompt, `data['value']` *(string, system prompt)*
 
+- `POST_PROMPT_ASYNC` - after preparing a system prompt, just before request in async thread, `data['value']` *(string, system prompt)*
+
+- `POST_PROMPT_END` - after preparing a system prompt, just before request in async thread, at the very end `data['value']` *(string, system prompt)*
+
 - `PRE_PROMPT` - before preparing a system prompt, `data['value']` *(string, system prompt)*
 
 - `SYSTEM_PROMPT` - when preparing a system prompt, `data['value']` *(string, system prompt)*
 
+- `TOOL_OUTPUT_RENDER` - when rendering extra content from tools from plugins, `data['content']` *(string, content)*
+
 - `UI_ATTACHMENTS` - when the attachment upload elements are rendered, `data['value']` *(bool, show True/False)*
 
 - `UI_VISION` - when the vision elements are rendered, `data['value']` *(bool, show True/False)*
@@ -3935,6 +3963,19 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.4.38 (2024-12-08)**
+
+- Added the ability to select a style for chat display between: Blocks, ChatGPT-like, and ChatGPT-like Wide. New option in the menu: Config -> Theme -> Style...
+- Added configuration options for audio input in Settings -> Audio -> Audio Input Device, Channels, and Sampling rate.
+
+**2.4.37 (2024-11-30)**
+
+- The `Query only` mode in `Uploaded` tab has been renamed to `RAG`.
+- New options have been added under `Settings -> Files and Attachments`: 
+  - `Use history in RAG query`: When enabled, the content of the entire conversation will be used when preparing a query if the mode is set to RAG or Summary.
+  - `RAG limit`: This option is applicable only if 'Use history in RAG query' is enabled. It specifies the limit on how many recent entries in the conversation will be used when generating a query for RAG. A value of 0 indicates no limit.
+- Cache: dynamic parts of the system prompt (from plugins) have been moved to the very end of the prompt stack to enable the use of prompt cache mechanisms in OpenAI.
+
 **2.4.36 (2024-11-28)**
 
 - Added a new command-line argument: --workdir="/path/to/workdir" to explicitly set the current working directory.
@@ -3968,7 +4009,7 @@ may consume additional tokens that are not displayed in the main window.
 
 - Added an option checkbox `Auto-index on upload` in the `Attachments` tab: 
 
-**Tip:** To use the `Query only` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `Query only` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
+**Tip:** To use the `RAG` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `RAG` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
 
 - Added context menu options in `Uploaded attachments` tab: `Open`, `Open Source directory` and `Open Storage directory`.
 

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/README.md

@@ -2,7 +2,7 @@
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.4.36** | build: **2024.11.28** | Python: **>=3.10, <3.12**
+Release: **2.4.38** | build: **2024.12.08** | Python: **>=3.10, <3.12**
 
 > Official website: https://pygpt.net | Documentation: https://pygpt.readthedocs.io
 > 
@@ -603,11 +603,11 @@ Built-in file loaders:
 - Webpages (crawling any webpage content)
 - YouTube (transcriptions)
 
-You can configure data loaders in `Settings / LlamaIndex / Data Loaders` by providing list of keyword arguments for specified loaders.
+You can configure data loaders in `Settings / Indexes (LlamaIndex) / Data Loaders` by providing list of keyword arguments for specified loaders.
 You can also develop and provide your own custom loader and register it within the application.
 
 LlamaIndex is also integrated with context database - you can use data from database (your context history) as additional context in discussion. 
-Options for indexing existing context history or enabling real-time indexing new ones (from database) are available in `Settings / LlamaIndex` section.
+Options for indexing existing context history or enabling real-time indexing new ones (from database) are available in `Settings / Indexes (LlamaIndex)` section.
 
 **WARNING:** remember that when indexing content, API calls to the embedding model are used. Each indexing consumes additional tokens. Always control the number of tokens used on the OpenAI page.
 
@@ -669,7 +669,7 @@ You can set the limit of steps in such a loop by going to `Settings -> Agents an
 
 You can change the prompt used for evaluating the response in `Settings -> Prompts -> Agent: evaluation prompt in loop`. Here, you can adjust it to suit your needs, for example, by defining more or less critical feedback for the responses received.
 
-##  Agent (Legacy, Autonomous) 
+##  Agent (Autonomous) 
 
 This is an older version of the Agent mode, still available as legacy. However, it is recommended to use the newer mode: `Agent (LlamaIndex)`.
 
@@ -817,11 +817,13 @@ The content from the uploaded attachments will be used in the current conversati
 
 - `Full context`: Provides best results. This mode attaches the entire content of the read file to the user's prompt. This process happens in the background and may require a large number of tokens if you uploaded extensive content.
 
-- `Query only`: The indexed attachment will only be queried in real-time using LlamaIndex. This operation does not require any additional tokens, but it may not provide access to the full content of the file 1:1.
+- `RAG`: The indexed attachment will only be queried in real-time using LlamaIndex. This operation does not require any additional tokens, but it may not provide access to the full content of the file 1:1.
 
 - `Summary`: When queried, an additional query will be generated in the background and executed by a separate model to summarize the content of the attachment and return the required information to the main model. You can change the model used for summarization in the settings under the `Files and attachments` section.
 
-**Important**: When using `Full context` mode, the entire content of the file is included in the prompt, which can result in high token usage each time. If you want to reduce the number of tokens used, instead use the `Query only` option, which will only query the indexed attachment in the vector database to provide additional context.
+In the `RAG` and `Summary` mode, you can enable an additional setting by going to `Settings -> Files and attachments -> Use history in RAG query`. This allows for better preparation of queries for RAG. When this option is turned on, the entire conversation context is considered, rather than just the user's last query. This allows for better searching of the index for additional context. In the `RAG limit` option, you can set a limit on how many recent entries in a discussion should be considered (`0 = no limit, default: 3`).
+
+**Important**: When using `Full context` mode, the entire content of the file is included in the prompt, which can result in high token usage each time. If you want to reduce the number of tokens used, instead use the `RAG` option, which will only query the indexed attachment in the vector database to provide additional context.
 
 **Images as Additional Context**
 
@@ -829,7 +831,7 @@ Files such as jpg, png, and similar images are a special case. By default, image
 
 **Uploading larger files and auto-index**
 
-To use the `Query only` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `Query only` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
+To use the `RAG` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `RAG` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
 
 ## Downloading files
 
@@ -2678,6 +2680,8 @@ Config -> Settings...
 
 - `Zoom`: Adjusts the zoom in chat window (web render view). `WebEngine / Chromium` render mode only.
 
+- `Style (chat)`: Chat style (Blocks, or ChatGPT-like, or ChatGPT-like Wide. `WebEngine / Chromium` render mode only.
+
 - `Code syntax highlight`: Syntax highlight theme in code blocks. `WebEngine / Chromium` render mode only.
 
 - `Font Size (chat window)`: Adjusts the font size in the chat window (plain-text) and notepads.
@@ -2700,8 +2704,6 @@ Config -> Settings...
 
 - `Use theme colors in chat window`: Use color theme in chat window, Default: True.
 
-- `Disable markdown formatting in output`: Enables plain-text display in output window, Default: False.
-
 **Files and attachments**
 
 - `Store attachments in the workdir upload directory`: Enable to store a local copy of uploaded attachments for future use. Default: True
@@ -2710,6 +2712,16 @@ Config -> Settings...
 
 - `Directory for file downloads`: Subdirectory for downloaded files, e.g. in Assistants mode, inside "data". Default: "download"
 
+- `Verbose mode`: Enabled verbose mode when using attachment as additional context.
+
+- `Model for querying index`: Model to use for preparing query and querying the index when the RAG option is selected.
+
+- `Model for attachment content summary`: Model to use when generating a summary for the content of a file when the Summary option is selected.
+
+- `Use history in RAG query`: When enabled, the content of the entire conversation will be used when preparing a query if mode is RAG or Summary.
+
+- `RAG limit`: Only if the option `Use history in RAG query` is enabled. Specify the limit of how many recent entries in the conversation will be used when generating a query for RAG. 0 = no limit.
+
 **Context**
 
 - `Context Threshold`: Sets the number of tokens reserved for the model to respond to the next prompt.
@@ -2802,14 +2814,22 @@ Config -> Settings...
 
 **Vision**
 
+- `Vision: Camera Input Device`: Video capture camera index (index of the camera, default: 0).
+
 - `Vision: Camera capture width (px)`: Video capture resolution (width).
 
 - `Vision: Camera capture height (px)`: Video capture resolution (height).
 
-- `Vision: Camera IDX (number)`: Video capture camera index (number of camera).
-
 - `Vision: Image capture quality`: Video capture image JPEG quality (%).
 
+**Audio**
+
+- `Audio Input Device`: Selects the audio device for Microphone input.
+
+- `Channels`: Input channels, default: 1
+
+- `Sampling Rate`: Sampling rate, default: 44100
+
 **Indexes (LlamaIndex)**
 
 - `Indexes`: List of created indexes.
@@ -3247,7 +3267,7 @@ If you want to only query index (without chat) you can enable `Query index only
 
 You can create a custom vector store provider or data loader for your data and develop a custom launcher for the application. 
 
-See the section `Extending PyGPT / Adding custom Vector Store provider` for more details.
+See the section `Extending PyGPT / Adding a custom Vector Store provider` for more details.
 
 # Updates
 
@@ -3545,6 +3565,8 @@ Syntax: `event name` - triggered on, `event data` *(data type)*:
 
 - `AI_NAME` - when preparing an AI name, `data['value']` *(string, name of the AI assistant)*
 
+- `AGENT_PROMPT` - on agent prompt in eval mode, `data['value']` *(string, prompt)*
+
 - `AUDIO_INPUT_RECORD_START` - start audio input recording
 
 - `AUDIO_INPUT_RECORD_STOP` -  stop audio input recording
@@ -3603,10 +3625,16 @@ Syntax: `event name` - triggered on, `event data` *(data type)*:
 
 - `POST_PROMPT` - after preparing a system prompt, `data['value']` *(string, system prompt)*
 
+- `POST_PROMPT_ASYNC` - after preparing a system prompt, just before request in async thread, `data['value']` *(string, system prompt)*
+
+- `POST_PROMPT_END` - after preparing a system prompt, just before request in async thread, at the very end `data['value']` *(string, system prompt)*
+
 - `PRE_PROMPT` - before preparing a system prompt, `data['value']` *(string, system prompt)*
 
 - `SYSTEM_PROMPT` - when preparing a system prompt, `data['value']` *(string, system prompt)*
 
+- `TOOL_OUTPUT_RENDER` - when rendering extra content from tools from plugins, `data['content']` *(string, content)*
+
 - `UI_ATTACHMENTS` - when the attachment upload elements are rendered, `data['value']` *(bool, show True/False)*
 
 - `UI_VISION` - when the vision elements are rendered, `data['value']` *(bool, show True/False)*
@@ -3845,6 +3873,19 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.4.38 (2024-12-08)**
+
+- Added the ability to select a style for chat display between: Blocks, ChatGPT-like, and ChatGPT-like Wide. New option in the menu: Config -> Theme -> Style...
+- Added configuration options for audio input in Settings -> Audio -> Audio Input Device, Channels, and Sampling rate.
+
+**2.4.37 (2024-11-30)**
+
+- The `Query only` mode in `Uploaded` tab has been renamed to `RAG`.
+- New options have been added under `Settings -> Files and Attachments`: 
+  - `Use history in RAG query`: When enabled, the content of the entire conversation will be used when preparing a query if the mode is set to RAG or Summary.
+  - `RAG limit`: This option is applicable only if 'Use history in RAG query' is enabled. It specifies the limit on how many recent entries in the conversation will be used when generating a query for RAG. A value of 0 indicates no limit.
+- Cache: dynamic parts of the system prompt (from plugins) have been moved to the very end of the prompt stack to enable the use of prompt cache mechanisms in OpenAI.
+
 **2.4.36 (2024-11-28)**
 
 - Added a new command-line argument: --workdir="/path/to/workdir" to explicitly set the current working directory.
@@ -3878,7 +3919,7 @@ may consume additional tokens that are not displayed in the main window.
 
 - Added an option checkbox `Auto-index on upload` in the `Attachments` tab: 
 
-**Tip:** To use the `Query only` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `Query only` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
+**Tip:** To use the `RAG` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `RAG` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
 
 - Added context menu options in `Uploaded attachments` tab: `Open`, `Open Source directory` and `Open Storage directory`.
 

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pygpt-net"
-version = "2.4.36-post1"
+version = "2.4.38"
 description = "Desktop AI Assistant powered by models: OpenAI o1, GPT-4o, GPT-4, GPT-4 Vision, GPT-3.5, DALL-E 3, Llama 3, Mistral, Gemini, Claude, Bielik, and other models supported by Langchain, Llama Index, and Ollama. Features include chatbot, text completion, image generation, vision analysis, speech-to-text, internet access, file handling, command execution and more."
 authors = ["Marcin Szczyglinski <info@pygpt.net>"]
 license = "MIT"

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/src/pygpt_net/CHANGELOG.txt

@@ -1,3 +1,16 @@
+2.4.38 (2024-12-08)
+
+- Added the ability to select a style for chat display between: Blocks, ChatGPT-like, and ChatGPT-like Wide. New option in the menu: Config -> Theme -> Style...
+- Added configuration options for audio input in Settings -> Audio -> Audio Input Device, Channels, and Sampling rate.
+
+2.4.37 (2024-11-30)
+
+- The `Query only` mode in `Uploaded` tab has been renamed to `RAG`.
+- New options have been added under `Settings -> Files and Attachments`: 
+  - `Use history in RAG query`: When enabled, the content of the entire conversation will be used when preparing a query if the mode is set to RAG or Summary.
+  - `RAG limit`: This option is applicable only if 'Use history in RAG query' is enabled. It specifies the limit on how many recent entries in the conversation will be used when generating a query for RAG. A value of 0 indicates no limit.
+- Cache: dynamic parts of the system prompt (from plugins) have been moved to the very end of the prompt stack to enable the use of prompt cache mechanisms in OpenAI.
+
 2.4.36 (2024-11-28)
 
 - Added a new command-line argument: --workdir="/path/to/workdir" to explicitly set the current working directory.
@@ -31,7 +44,7 @@
 
 - Added an option checkbox `Auto-index on upload` in the `Attachments` tab: 
 
-Tip: To use the `Query only` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `Query only` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
+Tip: To use the `RAG` mode, the file must be indexed in the vector database. This occurs automatically at the time of upload if the `Auto-index on upload` option in the `Attachments` tab is enabled. When uploading large files, such indexing might take a while - therefore, if you are using the `Full context` option, which does not use the index, you can disable the `Auto-index` option to speed up the upload of the attachment. In this case, it will only be indexed when the `RAG` option is called for the first time, and until then, attachment will be available in the form of `Full context` and `Summary`.
 
 - Added context menu options in `Uploaded attachments` tab: `Open`, `Open Source directory` and `Open Storage directory`.
 

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/src/pygpt_net/__init__.py

@@ -6,15 +6,15 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.11.28 01:00:00                  #
+# Updated Date: 2024.12.08 01:00:00                  #
 # ================================================== #
 
 __author__ = "Marcin Szczygliński"
 __copyright__ = "Copyright 2024, Marcin Szczygliński"
 __credits__ = ["Marcin Szczygliński"]
 __license__ = "MIT"
-__version__ = "2.4.36"
-__build__ = "2024.11.28"
+__version__ = "2.4.38"
+__build__ = "2024.12.08"
 __maintainer__ = "Marcin Szczygliński"
 __github__ = "https://github.com/szczyglis-dev/py-gpt"
 __website__ = "https://pygpt.net"

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/src/pygpt_net/controller/chat/attachment.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.11.26 04:00:00                  #
+# Updated Date: 2024.11.29 23:00:00                  #
 # ================================================== #
 
 import os
@@ -261,26 +261,22 @@ class Attachment(QObject):
         """
         return self.mode
 
-    def get_context(self, ctx: CtxItem) -> str:
+    def get_context(self, ctx: CtxItem, history: list) -> str:
         """
         Get additional context for attachment
 
         :param ctx: CtxItem instance
+        :param history Context items (history)
         :return: Additional context
         """
-        content = ""
-        meta = ctx.meta
         if self.mode != self.MODE_DISABLED:
             if self.is_verbose():
                 print("\nPreparing additional context...\nContext Mode: {}".format(self.mode))
 
-        self.window.core.attachments.context.reset()
-        if self.mode == self.MODE_FULL_CONTEXT:
-            content = self.get_full_context(ctx)
-        elif self.mode == self.MODE_QUERY_CONTEXT:
-            content = self.get_query_context(meta, str(ctx.input))
-        elif self.mode == self.MODE_QUERY_CONTEXT_SUMMARY:
-            content = self.get_context_summary(ctx)
+        self.window.core.attachments.context.reset()  # reset used files and urls
+
+        # get additional context from attachments
+        content = self.window.core.attachments.context.get_context(self.mode, ctx, history)
 
         # append used files and urls to context
         files = self.window.core.attachments.context.get_used_files()
@@ -296,34 +292,6 @@ class Attachment(QObject):
             return "====================================\nADDITIONAL CONTEXT FROM ATTACHMENT(s): {}".format(content)
         return ""
 
-    def get_full_context(self, ctx: CtxItem) -> str:
-        """
-        Get full context for attachment
-
-        :param ctx: CtxItem instance
-        :return: Full context
-        """
-        return self.window.core.attachments.context.get_context_text(ctx, filename=True)
-
-    def get_query_context(self, meta: CtxMeta, query: str) -> str:
-        """
-        Get query context for attachment
-
-        :param meta: CtxMeta instance
-        :param query: Query string
-        :return: Query context
-        """
-        return self.window.core.attachments.context.query_context(meta, query)
-
-    def get_context_summary(self, ctx: CtxItem) -> str:
-        """
-        Get context summary
-
-        :param ctx: CtxItem instance
-        :return: Context summary
-        """
-        return self.window.core.attachments.context.summary_context(ctx, ctx.input)
-
     def get_uploaded_attachments(self, meta: CtxMeta) -> list:
         """
         Get uploaded attachments for meta

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/src/pygpt_net/controller/config/placeholder.py

@@ -86,6 +86,8 @@ class Placeholder:
             return self.get_agent_providers()
         elif id == "syntax_styles":
             return self.get_syntax_styles()
+        elif id == "styles":
+            return self.get_styles()
         elif id == "idx":
             return self.get_idx()
         elif id == "keys":
@@ -98,9 +100,23 @@ class Placeholder:
             return self.get_speech_synthesis_actions()
         elif id == "voice_control_actions":
             return self.get_voice_control_actions()
+        elif id == "audio_input_devices":
+            return self.get_audio_input_devices()
         else:
             return []
 
+    def get_audio_input_devices(self) -> list:
+        """
+        Get audio input devices list
+
+        :return: placeholders list
+        """
+        devices = self.window.core.audio.get_input_devices()
+        data = []
+        for device in devices:
+            data.append({str(device[0]): device[1]})
+        return data
+
     def get_langchain_providers(self) -> list:
         """
         Get Langchain LLM provider placeholders list
@@ -276,6 +292,19 @@ class Placeholder:
             data.append({id: id})
         return data
 
+    def get_styles(self) -> list:
+        """
+        Get styles list
+
+        :return: placeholders list
+        """
+        styles = self.window.controller.theme.common.get_styles_list()
+        styles.sort()
+        data = []
+        for id in styles:
+            data.append({id: id})
+        return data
+
     def get_keys(self) -> list:
         """
         Get keys

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/src/pygpt_net/controller/lang/mapping.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.11.23 00:00:00                  #
+# Updated Date: 2024.12.07 21:00:00                  #
 # ================================================== #
 
 from pygpt_net.utils import trans
@@ -357,7 +357,7 @@ class Mapping:
         menu_text['config.profile.new'] = 'menu.config.profile.new'
         menu_text['config.save'] = 'menu.config.save'
         menu_text['theme.tooltips'] = 'menu.theme.tooltips'
-        menu_text['theme.blocks'] = 'menu.theme.blocks'
+        menu_text['theme.style'] = 'menu.theme.style'
         menu_text['theme.settings'] = 'menu.theme.settings'
         menu_text['plugins.presets.new'] = 'menu.plugins.presets.new'
         menu_text['plugins.presets.edit'] = 'menu.plugins.presets.edit'

{pygpt_net-2.4.36.post1 → pygpt_net-2.4.38}/src/pygpt_net/controller/settings/editor.py

@@ -63,6 +63,7 @@ class Editor:
         self.window.ui.add_hook("update.config.debug", self.hook_update)
         self.window.ui.add_hook("update.config.notepad.num", self.hook_update)
         self.window.ui.add_hook("update.config.render.code_syntax", self.hook_update)
+        self.window.ui.add_hook("update.config.theme.style", self.hook_update)
         # self.window.ui.add_hook("llama.idx.storage", self.hook_update)  # vector store update
         # self.window.ui.add_hook("update.config.llama.idx.list", self.hook_update)
 
@@ -163,6 +164,11 @@ class Editor:
             value = self.window.core.config.get('render.code_syntax')
             self.window.controller.theme.toggle_syntax(value, update_menu=True)
 
+        # style
+        if self.config_changed('theme.style'):
+            value = self.window.core.config.get('theme.style')
+            self.window.controller.theme.toggle_style(value)
+
         # convert lists
         if self.config_changed('ctx.convert_lists'):
             self.window.controller.ctx.refresh()